With the development of intelligent terminal devices, a screen capture function has become a basic configuration of the terminal device. However, up to the present, for most terminal devices, screen capture is implemented by using a combination of hardware buttons, and only an image displayed on an entire screen can be captured, which lacks content pertinence; a user usually needs to perform further processing such as clipping on a captured image, where the operation is tedious and complex. For some latest terminal devices, screen capture for a specific region can be implemented by recognizing a swipe track of a stylus on a screen. However, the screen capture for a specific region can only be implemented with the help of the stylus; and therefore, a screen capture operation is tedious and complex, and user experience is yet to be further improved.